Sir Alektron . . . . .
Au contraire . . . . . .messr . . . . it's just that there is lots of conflicting info on those two unit schematics and the amassing and amending of that info.
With my using your two best board shots of max relevance, and the amended schema . . . . we now continue.
Reviewing . . . .
The power supply passage of the + and - 15VDC supplies to the main board seem to be fine.
There is one interlocked voltage feature between the two boards in the respect of the +15 from the power supply needing to feed a sub regulated reference voltage to pin 5 of the IC34B section comparator.
DC voltage meter across C72 electrolytic, and see how close to 5.1 VDC that you are reading.
When CN107 gets interconnected to the PS board the VCC at pins 13 and 14 plummets down to hundreds of millivolts.
Your testing of Q8 power switching transistor reveals no dead short from C to E.
The diode that creates the 5 VDC Dee supply is being the D23 Schottky and is marked up on the foil pattern side as a fine line YELLOW rectangle, with it being a flat pack surface mount.
There is a black on black situation where there is being a jet black stripe of adhesive holding the flat pack in place during factory flow soldering bath operations.
Confirm that the adhesive is just being a bit in excess and being sloppy looking and that the diode case has
NOT been hot and cracked and pooched its case from an overload and the heat thereby being produced.
A deviavce . . .
Evaluating the replacement of the super cap and its marked fine YELLOW circles, you commented on the hardmess of the two solder joints.
I am thinking that this boards birth goes back before the manufacturing transition to lead free solder.
What solder are you using now ? 60/40 rosin cored or the lead free type. Looks like you didn't do a pre-prep of applying a microdrop of liquid rosin flux to the joints before soldering.
That, along with a touch of rosin core 60/40 fresh solder and a well tinned tiplet should have resulted in a very shiny joint (example D8) instead of the resultant mottled grey.
I need Q8 transistor removed from the board, using that same 3 solder ball technique and rapidly walking the soldering iron tiplet between the 3 balls until Q8 leads reach solder release time.
Then the unit needs DC voltage monitoring of a connected CN107 to its mate at the pins 12 and 13, powering up for ONLY the time that is required for a voltage to show up on the meter . . . then power down.
Thassssit . . . . .
RELEVANT REFERENCING . . . . .
73's de Edd
.....